KerasObjectDetector

KerasObjectDetector copied to clipboard

mike2ox

[Preprocessing] Object Detection Tutorial(used by Keras API)

2

- [x] Explain Obeject Detection concept - [ ] Explain Obeject Detection Model([Faster R-CNN](https://arxiv.org/pdf/1506.01497.pdf)) ➡️ DateSet([Deepfashion2](https://arxiv.org/pdf/1901.07973.pdf)) - [ ] Coding Tutorial Code ➡️ Explain Code

mike2ox

The error messages : ``` serWarning: File /home/yolk/VOCdevkit/yolk_voc_train_val2012_tar is downloaded but the content hash does not match. The repo may be outdated or download may be incomplete. If the "repo_url"...

minus31

Make argument parser that can be used for various backbone model including RetinaNet, YOLO, and SSD.

ahracho

- [ ] Create and add yolo annotation files - COCO / VOC Dataset - [ ] Replace train.txt to each annotation file names according to dataset type

simba328

Create custom shapes dataset - Checklist - [x] Shape datasets Modularization - [x] make bounding box - [ ] data generator(dataloader) - Reference + [matterport's Mask-RCNN shapes dataset](https://github.com/matterport/Mask_RCNN/blob/master/mrcnn/model.py)

SooDevv

Make a higher API for wrapping * Select model type(SSD, RetinaNet, YOLOv3) * Select DataLoader

MijeongJeon

[modeling] Add SSD model on dev branch

4

# Add SSD model on dev branch Github repo: [ssd_keras](https://github.com/pierluigiferrari/ssd_keras) by @pierluigiferrari

hngskj

RetinaNet Code Review

1

https://github.com/fizyr/keras-retinanet

visionNoob

I wanna put "pre-processing module" inside of our model

1

mean을 빼고 std로 나누는 등 전처리 연산을 모델 레벨에서 처리하면 좋을 것 같은데.. see also https://stackoverflow.com/questions/44835223/adding-a-preprocessing-layer-to-keras-model-and-setting-tensor-values

visionNoob